MiniMax M3 428B β MI325X vs MI355X Performance per Dollar
Cost per million tokens of MI325X (AMD CDNA 3) versus MI355X (AMD CDNA 4) on MiniMax M3 428B. Large-hyperscaler-volume ownership TCO normalized by total tokens β performance per dollar across LLM workloads. Pick the more cost-efficient SKU at every target interactivity level. Use the chart controls below to switch sequences, precisions, and metrics β same interactions as the main inference chart.
AgentX replays real coding-agent sessions rather than fixed-length prompts, so context grows turn over turn and most of each request is served from cache instead of being recomputed. That turns the comparison into a systems question: KV transfer between nodes, prefix-aware routing, and cache capacity all move the curve alongside raw chip throughput. Fixed-sequence workloads stay the clean baseline for kernel and silicon performance, so the two scenarios answer different questions about the same hardware. Learn more about AgentX β
MI355X costs $0.01 per million tokens at 150 tok/s/user on MiniMax M3 428B; we have no MI325X benchmark data at this exact target.
At 257 tok/s/user on MiniMax M3 428B, MI355X comes in at $0.02 per million tokens. MI325X hasn't been benchmarked at this operating point.
Only MI355X has cost data at 363 tok/s/user on MiniMax M3 428B β $0.04 per million tokens. MI325X is unmeasured at this target. (Numbers reflect the default agentic-traces Β· fp4 selection for this URL β table and chart below update if you change sequence, precision, or model in the controls.)
Chip pricing (owning hyperscaler): MI325X $1.10/chip/hr Β· MI355X $1.50/chip/hr. Source: SemiAnalysis Market July 2026 Pricing Surveys & AI Cloud TCO Model.
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Dollar per Million Tokens | MI325X:βMI355X:$0.010 | MI325X:βMI355X:$0.021 | MI325X:βMI355X:$0.044 |
| Concurrency | MI325X:βMI355X:~23 | MI325X:βMI355X:~11 | MI325X:βMI355X:~4 |
Inference Performance
Agentic inference metrics from the AgentX scenario and fixed-sequence inference metrics across models, hardware configurations, and serving parameters.